To track their blood sugar levels, patients with diabetes typically prick their fingers at least three times a day and feed blood samples into glucometers. It’s a tedious and sometimes painful process, and a patient will often need to run a second test due to “insufficient blood” in the first sample. Now, researchers at Baylor University, in Waco, TX, have engineered a thumb-pad sensor that measures glucose levels via electromagnetic waves–no finger pricking required.
“There are many patients that don’t monitor because of the pain of monitoring,” says John Buse, president of the American Diabetes Association. “So there’s certainly the potential to improve the lives of people with diabetes.”
According to Randall Jean, associate professor of electrical and computer engineering at Baylor, the prototype of the new device matches the performance of conventional glucometers.
“It is accurate enough for people to make decisions about whether or not to inject insulin,” says Jean. “That’s really the target. It’s not to measure glucose within one ppb [part per billion] but to produce an instrument that patients can use to make decisions about externally controlling blood sugar.”
The U.S. Food and Drug Administration has approved only one noninvasive glucose monitor, called the GlucoWatch Biographer. Designed by Cygnus, of Redwood City, CA, the device is a wristwatch that uses an electric current to pull small amounts of fluid through the skin without pricking it. A sensor analyzes the fluid for glucose. However, 50 percent of patients who used the watch experienced skin irritation and sores, and the product was discontinued in 2007.
Jean says that the sensor he and his colleagues are developing will be “truly noninvasive” and will not require that any fluid–blood or otherwise–pass through the skin. The sensor itself is a small, spiral-shaped microwave circuit, which acts as a transmission line and emits electromagnetic waves. When a person places her thumb on the spiral, the electrical properties of her thumb change how energy passes through the circuit. Jean and his colleagues measure this change, and in early trials, they seem to have found patterns that correspond to variations in glucose levels.
“The energy does not specifically respond to glucose; it responds to the aggregate effect of blood, muscle, fat, skin, and glucose,” says Jean. “What we’re hoping is that over a broad enough frequency range, the individual components have unique signatures that allow us to extract the glucose.”
The sensor is still in the early stages of development, and Jean has so far tested the prototype on five volunteers in 15 separate trials. The researchers made plastic molds of each subject’s thumb, and they fabricated plastic guides to ensure that the subjects placed their thumbs on the sensors in exactly the right position. Jean also added a pressure gauge to tell the subjects how hard to press down in order to get a successful read. In each trial, volunteers placed their thumbs on the sensors, and researchers took 10 separate readings. Subjects also performed finger-prick tests, drawing blood and using traditional glucometers.
Researchers entered data from both methods into a computer program and looked for patterns within the electromagnetic data that corresponded to the glucose readings from the blood samples.
Although the early results are promising, it remains to be seen how broadly they can be generalized. “We’re still working on verifying that the calibrations are truly robust,” says Jean. “In other words, the data looks good for the people we’ve had a lot of experience with, but now we have to make sure that if a new thumb comes along, it works on that one.”
What’s more, Jean’s sample pool tended to exhibit glucose levels within the normal range. To verify the sensor’s accuracy, the team needs to test it on volunteers with varying glucose levels. In the next few months, Jean plans to test the sensor on patients at Scott and White Hospital, in Temple, TX, whose glucose readings may be “all over the map.”
“If [a monitor] could be developed, it would be enormously promising because it’s not just noninvasive but could give continuous data,” says Howard Wolpert, director of the Insulin Pump Program at the Joslin Diabetes Center, which is based in Boston. “That’s what people are interested in, because with devices today, you’re only looking at intermittent time points, and glucose fluctuations can be quite dramatic.”
Jean says that while his ultimate goal is to design an accurate sensor cheap enough for patients to carry around with them, he expects that one of the first early uses of the technology will be as screening devices at local drugstores, much like the large commercial monitors that take blood-pressure and heart-rate readings.
“It could provide a useful service for someone who didn’t know they’re diabetic, and you could say, ‘Your blood sugar is kind of high. You should go to the doctor,’” says Jean.
The dark secret behind those cute AI-generated animal images
Google Brain has revealed its own image-making AI, called Imagen. But don't expect to see anything that isn't wholesome.
The hype around DeepMind’s new AI model misses what’s actually cool about it
Some worry that the chatter about these tools is doing the whole field a disservice.
The walls are closing in on Clearview AI
The controversial face recognition company was just fined $10 million for scraping UK faces from the web. That might not be the end of it.
This horse-riding astronaut is a milestone in AI’s journey to make sense of the world
OpenAI’s latest picture-making AI is amazing—but raises questions about what we mean by intelligence.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.